search

joslefaure / HIT

Official Implementation of our WACV2023 paper: “Holistic Interaction Transformer Network for Action Detection”
https://arxiv.org/abs/2210.12686
55 stars 9 forks source link

Backward Error #24

Closed Tsunehiko closed 1 year ago

Tsunehiko commented 1 year ago

I tried running python train_net.py --config-file "config_files/hitnet.yaml", but I encountered the following error. Could you advise on how to resolve this? I installed PyTorch 1.10.0. Thank you.

RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [8, 1024]], which is output 0 of ReluBackward0, is at version 1; expected version 0 instead. Hint: the backtrace further above shows the operation that failed to compute its gradient. The variable in question was changed in there or anywhere later. Good luck!

joslefaure commented 1 year ago

The code was tested with torch 1.71

Tsunehiko commented 1 year ago

Thanks, I'll try it